Design of A Planar Parallel Robot For Optimal Workspace and Dexterity

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/221915447
Design of a Planar Parallel Robot for Optimal Workspace and Dexterity
Article in International Journal of Advanced Robotic Systems · September 2011

DOI: 10.5772/45693 · Source: InTech
CITATIONS READS
21 641
1 author:
Ming Z Huang
University of San Diego
35 PUBLICATIONS 233 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Developing Changemaking Engineers View project
All content following this page was uploaded by Ming Z Huang on 28 January 2014.
The user has requested enhancement of the downloaded file.

Design of a Planar Parallel Robot for
Optimal Workspace and Dexterity
Regular Paper
Ming Z. Huang*

Department of Engineering, University of San Diego, USA
*Corresponding author E-mail: minghuang@sandiego.edu
Received 12 May 2011; Accepted 19 Aug 2011

Abstract Parallel robots exhibit salient merits over their effector can reach with at least one orientation, whereas
serial counterparts in applications where both accuracy the singularities are special geometric configurations
and dynamic response are required. However, due to the inherent to the workspace at which the manipulator can
strong dependence of geometric parameters and their lose–if actuated in series, or gain–if actuated in parallel,
performances, the corresponding design problems for the one or more degrees of freedom.
parallel robots are much more complex and the adequacy
and effectiveness of the design method become more Being able to avoid the singularities is more critical in
critical. In this paper, a study in the design optimization practice for parallel manipulators than their serial
for a class of planar parallel robots is presented. The counterparts due to potential control uncertainty
robots feature an in‐parallel structure with two degrees of associated with the unwanted, albeit instantaneous,
freedom. Dimension synthesis is performed through mobility even with all actuators locked. Such uncertainty
maximization of two key performance characteristics, also manifests itself in poor force transmission
addressing not only workspace but also dexterity of the performance, in that the manipulator cannot effectively
robots under consideration. Optimal designs are attained resist or apply forces or torques at the end‐effector in
using both parametric study and simplex algorithm. certain direction. Ideally a robot should have as large a
Results are shown by way of computer simulations aided workspace and as small a density in singularity as
with graphic visualizations. possible, both being functions of robot geometries.
Attaining such a kinematic property requires a dimension
Keywords Optimal design, Parallel robots, Workspace, synthesis process through some optimization.
Dexterity

Many researchers have addressed the topics of
workspace and singularity of parallel manipulators. For
1. Introduction examples, the boundary of the dexterous workspace for a
3‐DOF planar parallel manipulator was determined by
In robot design as well as control coordination, geometric reasoning in [1]. Geometric algorithms were
workspace and singularity characteristics are two presented to obtain the maximal as well as the total
important geometric properties that should be orientation workspace for a similar type of manipulator
considered. The workspace is the region which the end‐ [2]. For a broader class of parallel manipulators, a study
176 Int J Adv Robotic Sy, 2011, Vol. 8, No. 4, 176-183 www.intechweb.org
www.intechopen.com
to classify the various shapes of workspace due to
changes in link length was reported in [3]. Representative
theoretical studies on the singularity of parallel
manipulators can be found with geometrical approach,
by use of the Grasmaan geometry [4], screw theory [5], or
using related line geometry [6]. And in [7] this author
reported a study of workspace and singularity
characteristics for two common types of 3‐DoF planar
parallel manipulators, in which the complete workspace
and the singularity inherent to the systems were
characterized using computer simulations based on a
simple direct search algorithm aided with graphic

visualization. Both the workspace and the singularity
Figure 1. The 5R planar parallel robot
related properties have been, individually or together, as

primary objectives in many optimal design studies for
2.1 Forward Position Kinematics
parallel robots. For examples the workspace was used

solely as the metric of optimization in [8],[9] whereas
The robot is driven by the two angles θ1 and θ2. Each set
indices of singularity, stiffness, and accuracy were used
of (θ1, θ2) corresponds to a scalar couple (X, Y) which
either in composite (e.g., as a weighted sum in [10],[11])
represents coordinates of the tool point Q. The following
or multi‐criteria form in optimal design [12].
derives the expressions of X and Y in terms of θ1 and θ2.

The coordinates of point Q can be expressed, from the
In this paper, a study in the design optimization for a
side O1O2BQ and the side O1APQ, as:
class of planar parallel robots is presented. The robots

feature an in‐parallel structure with two degrees of
� � � � �� 1 � ��
freedom. Dimension synthesis is performed through
�� (1)
optimization of two key geometric properties, namely

addressing not only workspace but also dexterity of the � � �� 1 � ��
robots under consideration. A convenient composite �� (2)
index, defined as the ratio of the total available
workspace to the density of singularity within that Isolating the cos3 and sin3 terms from the second
workspace, was developed in [7] to help inform the way equations above, then squaring and adding the two
in which both characteristics vary with respect to various resulting equations to eliminate 3, the following
geometric parameters. The proposed index, considered as expression is obtained:
a more proper representation of the overall dexterity of
the robot, is adopted as the metric for optimal design in �� (3)
this study. A solution to optimal design is arrived at and
validated by using both the parametric variations and the where
simplex algorithm independently. Results are shown by
way of computer simulations aided with graphic � � �� ; � � ��
visualizations.
1
2. System Model � � ��
�

The robot of interest is a two‐degree‐of‐freedom five‐bar Equation (3) is a linear trigonometric equation, which can
planar parallel robot (see Fig. 1). The robotic mechanism be solved by using the tan‐half angle transformation to
consists of a point P (ʹwristʹ) supported by two articulated give two closed form solutions of cos and sin in terms
arms. Each arm is composed of two links connected by of θ1 and θ2. Then by equations (1) and (2), the position
five revolute joints between the links (at points A and B, coordinates of tool point Q can be obtained. The forward
P) and to the ground (points O1 and O2). kinematic solution is necessary to help visualize the robot
corresponding to any given set of inputs θ1 and θ2. The
The link lengths of each two‐ link arm are L and m, two possible configurations corresponding to a given set
respectively. Point Q is the tool center point which is of input angles (θ1, θ2) are illustrated in Fig. 2.
located at m from the wrist point P along the upper right
arm.

www.intechopen.com
2.3 Velocity Kinematics Analysis

The kinematics equation relating the velocity of the tool
point and the input joint rates, is characterized by a linear
mapping matrix J, known as the Jacobian. The Jacobian is
required for the characterization of geometric
singularities of the system. This section specifies the
Jacobian for the planar robot under study.

To derive the Jacobian matrix, equations (6) and (11) are
differentiated, respectively:
��
�� (12)
Figure 2. Two configurations for the same inputs (θ1,θ2)
��
2.2 Inverse Kinematics Solution ��
�� (13)
The inverse kinematics solution specifies the expressions
of θ1 and θ2 in terms of the tool point coordinates (X, Y). where
The closed form expressions for θ1 and θ2 are derived � � 2��
next. Starting from the left branch, equations (1) and (2) is � ��
rewritten as:

� � � � �� 1 � �� (4) In order to express �� in terms of ��, ��, and �� , equations
(7) and (8) are differentiated, which lead to two
� � �� 1 � �� (5)
possibilities:
Summing the squares of (4) and (5) gives:
1 ��
2�� 2�� 1 � ��
��1 � �� 1 � ��
�� (6)
�1 ��
Equation (6) again is a linear trigonometric equation in θ2, ��
��1 � �� 1 � ��
which can be solved, similar to equation (3), to yield two
solutions of cosθ2 and sinθ2. From equations (4) and (5),
Upon substitution and arrangement, equations (12) and
the following expressions for θ4 can be obtained:
(13) can be written using the matrix form:
�� ⁄��1 � �� (7)
� ��
�� ⁄��1 � �� (8) �� .
��
Now considering equations for the right branch in That is,
equations (1) and (2), we have:
��
�� (9) � � � � � � ��
��
�� (10)
where
The sum of the squares of equations (9) and (10) gives:

�� (11) �� 2��

where �� 2��

� � 2�� 2��
�� 2��

� � ��
� � � � � � �� 2��
Equation (11) again yields two solutions for θ1. Thus �
� ��
there are four possible configurations for the robot to ��1 � ��
reach a given point (X, Y). Note that, because singularity
�
is a property of geometric configuration, the robot can be �� 2��
��1 � ��
at singularity in any one of the four configurations.
�� 2��

www.intechweb.org Ming Z. Huang: Design of a Planar Parallel Robot for Optimal Workspace and Dexterity 178
www.intechopen.com
And Fig. 3 shows the workspace of the robot obtained using
such an algorithm implemented using Matlab based on
݆௧ଵଵ ൌ Ͳ the inverse kinematics solution. One advantage of this

approach is that it allows for effective quantification of
݆௧ଵଶ ൌ ʹ‫ܮ‬ሾሺܻܿ‫ߠݏ݋‬ଶ ሻ െ ሺܺ െ ܾሻ‫ߠ݊݅ݏ‬ଶ ሿ
workspace area as well as singularities.
݆௧ଶଵ ൌ ʹ‫ܮ‬ሾሺܺ െ ߚ݉ܿ‫ߠݏ݋‬ସ ሻ‫ߠ݊݅ݏ‬ଵ െ ሺܻ െ ߚ݉‫ߠ݊݅ݏ‬ସ ሻܿ‫ߠݏ݋‬ଵ ሿ
3.2 Singularity Analysis
‫ߠݏ݋ܿܮ‬ଶ
݆௧ଵଶ ൌ ߙ Ǣ ݂݅‫ߠ݊݅ݏ‬ସ ൌ Ͳ
݉ሺͳ ൅ ߚሻܿ‫ߠݏ݋‬ସ
When the tool point Q is at a singular point of the

workspace, the robot is actually in a uncertainty
‫ߠ݊݅ݏܮ‬ଶ
ൌ ߙ Ǣ ݂݅‫ߠ݊݅ݏ‬ସ ് Ͳ configuration in which it is uncontrollable. It is necessary
݉ሺͳ ൅ ߚሻ‫ߠ݊݅ݏ‬ସ
to identify so as to avoid all such configurations in
Finally the Jacobian matrix J can be obtained as:
‫ ܬ‬ൌ ‫ܬ‬௉ିଵ Ǥ ‫ܬ‬௧ planning a trajectory for the robot. As such
characterization of singularity is just as important as that
3. Evaluations of Geometric Properties of workspace, and some optimization of singularity
should be taken into consideration in the design of robot.
Workspace and singularity are the two most important
properties that are of concern in geometric design of robot. There are several indexes that characterize the degree of
In the case for the 2D robot being considered, the former is singularity; the determinant‐based manipulability and
the collection of all the points reachable by the tool points the (reciprocal) condition number of the Jacobian, are the
and the latter corresponds to the configurations in which two commonly used ones. When the determinant of the
the robot becomes singular (gaining a transitory degree of Jacobian is equal to zero, the robot is at singularity.
freedom at uncertainty configuration). However, the magnitude of the determinant cannot be
used as an appropriate measure of the degree of ill‐
3.1 Workspace Analysis conditioning. To this end the better metric to use is the
reciprocal condition number of the Jacobian. The
The inverse kinematic solution is used in the studies for the condition number based index, icd, is bounded between 0
workspace of the robot. Indeed, given the coordinates (X, Y) and 1. Obviously, the conditioning index depends on the
of any point in the region, it can be determined if it belongs joints coordinates of the manipulator and therefore it is
to the workspace or not by verifying if at least one solution posture‐dependent. As each tool point is being identified
for θ1 and θ2 exists. The workspace can be evaluated by through the workspace analysis, the icd for corresponding
using a direct search algorithm which essentially ʹgridsʹ the Jacobian can be readily computed to characterize the
potential region in conjunction with checking the existence singularity‐degree of each point in the workspace.
of an inverse kinematics solution for the robot. Such an
algorithm has been used to study the workspace In this study, the robot is considered being at or near a
characteristics of a class of three degree of freedom robots singularity when icd < 0.01. A Matlab program was used
[7]. to generate a graph which gives the robot workspace and
its degree of singularity of each tool point as follows (see
Fig. 4):

Figure 4. Workspace and singularities of the 2 DoF robot

Figure 3. Workspace of the 2 DoF robot
www.intechopen.com
Thus the design optimization problem, specifically for the
5R planar robot considered here, can be stated as:

Find the link lengths b, L, m and βm that maximize an
objective function that quantifies the overall dexterity of the
robot.

In the evaluation of the objective function based on the
above dexterity index, the percentage of workspace is
obtained from the ratio of actual robot workspace area
relative to a reference area given by a circle with a radius
of (L+m+βm), which corresponds to the widest area the
robot could cover if it had only one arm.

The robot workspace area is obtained by associating a
small area to each point of the grid in a target window,
where 2Lmax is the size of the target window (x and y
ranging from ‐Lmax to +Lmax) and n is the discretization
Figure 5. Ambiguity of singular configuration due to multiplicity precision (the number of points in one axis). The
of solutions for the same tool point workspace area is computed by multiplying this small
area with the number of points for which the inverse
Recall that there are four possible robot configurations for kinematics solutions exist. The percentage of singularities
the same point of the workspace. Therefore, there are four is obtained from the ratio of area based on the collection
different Jacobians and thus four different icdʹs for each of those singular points (with icd less than 0.01) and the
point of the workspace. This fact complicates the area of the robot workspace.
characterization of singularity and, if only one of the
solutions is considered during the study, an incomplete 4.1 Optimization by Parametric Variations
or even mistaken conclusion could result. For example in
Fig. 5, the first of the four possible configurations is at The design problem of interest here is one of
singularity whereas the others are clearly not. multivariable nonlinear optimization type. A basic
approach to finding a solution is by means of parametric
4. Design Optimization variation, where a candidate solution vector is generated
by locating the optimal value for each parameter while
In robot design, an ideal geometry is one that has the keeping all others fixed, which candidate solution is then
widest workspace possible and the fewest singularities in used as the new basis for iteration by repeating the
its workspace. Indeed, from a motion planning point of aforementioned process to generate the next candidate.
view, a robot with a large workspace filled with many The process stops until convergence is attained. This
singular points, is just as undesirable as one that has a method is similar to the successive substitution method
small workspace. In this study, an optimal design of the 2 for root finding applications, which is tedious in its
DoF planar 5R robot is obtained for its link lengths by process but quite effective in finding a solution.
taking both the workspace and singularity properties into
consideration. This is realized through the use of a In this case a candidate value for each of the link
composite index relating the two properties as proposed parameters is identified numerically by characterizing
previously in [7]. participation of each parameter in the dexterity for which
a graph may be generated as a function of a varying
The composite index, named as the overall dexterity parameter while keeping all other parameters fixed; the
index, is calculated by taking the ratio of two metrics, value corresponding to the greatest dexterity is then
namely a percentage measure of the workspace relative to reported.
a specific reference (chosen by the designer) and a
percentage measure of the extent of singularity relative to The optimization process is demonstrated in the
the workspace. Since both properties of workspace and following with a discretization precision of 100 points in
singularity are functions of dimensions and geometric the X‐axis and the Y‐axis. All link lengths, namely b, L,
configuration of the robot, the design problem for a and m, referred to heretofore may be of any convenient
chosen robot configuration becomes one of finding the set unit as long as they are consistent, and  is a
of link dimensions that maximize the dexterity index. dimensionless parameter that sets the length of the end
effector.
www.intechopen.com
First Iteration ‐ To start the optimization process, a Notice that in Fig. 7, the dexterity values are abnormally
baseline design is first chosen as follows: b = 1, L=3, m = 2, high for L < 0.5b, which was found to be the natural effect
 = 0.3. A series of simulations were then conducted to of small values in the denominator due mainly to the fact
obtain the four graphs of parametric variations as shown that there is almost no singular point in the workspace.
in Fig. 6‐9. This configuration, however, cannot be considered as
optimal because the workspace is extremely limited and
disjointed (Fig. 10). Also it can be noted in Fig. 9 that the
dexterity jumps for b > 6 (or 2L), which can be attributed
to the phenomenon of workspace changing from one to
two regions, similar to that illustrated in Fig. 10.

Figure 6. Influence of β in the dexterity ‐ optimal at β = 0.05

Figure 9. Influence of b in the dexterity ‐ optimal at b = 1.8

From these graphs, the optimal values for the four design
parameters can be identified as: b = 1.8, L=1.9, m = 1.5,
 = 0.05.

Figure 7. Influence of L in the dexterity ‐ optimal at L = 1.9

Figure 10. An example of singularity‐free solution

Second Iteration ‐ Having found the first candidate
solution above, it is ʹsubstitutedʹ back to the optimization
process in which simulations are repeated to identify a
new set of optimal parameters as before. The results show
that the optimal values for  = 0.05 and L = 1.9 which
remain the same as before; however, the optimal values
Figure 8. Influence of m in the dexterity ‐ optimal at m = 1.5 for m and  did change (from 1.5 and 1.8) to be m = 2.2
and b = 1.34.
www.intechopen.com
Third Iteration ‐ The next step of the optimization is now
to determinate whether the optimal value of m depends
on b and vice versa. To do so, the evolution of the
dexterity with respect to b with m = 2.2 (last optimal value
found for m) is first generated. The maximum dexterity is
obtained for b = 1.407. This is in turn used to investigate
the variations of dexterity relative to m with b = 1.407,
which renders a new optimal value for m = 2.1. Then by
substituting this new m value and repeating the
simulations again, the new b value is found to be
convergent at b = 1.407. With these new values for m and
b, the optimal values for β and L have to be verified. The
evolutions of the dexterity depending on β on the one
hand and on L on the other using these values for b and m
demonstrate that the optimum for β and L is not affected.
Figure 12. Singularity characteristics of the optimized 2 dof robot
The Optimal Solution ‐ The optimal parameters of the
robot are: b = 1.407, L = 1.9, m = 2.1,  = 0.05. Fig. 11 shows In order to corroborate the result obtained previously ʹby
the optimal robot geometry and its workspace, and Fig. handʹ the optimization was carried out using the built‐in
12 gives the distribution of singularity. And the optimal Matlab function with a precision of 100 for the
design corresponds to the following numerical discretization.
properties:

 Workspace area : 40.9480 The optimization process was executed using the same
 Normalized area ratio : 0.7735 initial guess as before; namely b = 1, L= 3, m = 2,  = 0.3.
 Singularity percentage : 0.0444 The following optimum is returned by the fminsearch
 Dexterity : 17.4036 algorithm: b = 1.0545, L = 2.8806, m = 2.0228, = 0.3074,
which corresponds to a dexterity of 11.6828, a value
different from the optimum found previously. This shows
that if the previous result is correct, the direct search
simplex algorithm did not find the true optimum but
only a local optimum close to the initial guess; this is no
doubt a pitfall of the simplex algorithm.

To validate the result found previously ʹby hand,ʹ a
second iteration is executed using those values found by
parametric variations as a new set of initial guess. The
following optimum is obtained: b = 1.4052, L = 1.9017, m =
2.1007,  = 0.0501, corresponding to dexterity of 17.9350.
Comparing these to the values found by hand, it can be
concluded that both results corroborated with each other.

One may wonder whether if the number of discretization
Figure 11. Workspace of the optimized 2 DoF robot used will affect the results. To investigate the effect, a
different number for the discretization is used (n = 200,
4.2 Optimization by Simplex Algorithm namely 40,000 points). Again using the optimal values
found by hand as initial guess, the following optimum is
The foregoing optimization, being a multi‐variable, obtained: b = 1.4599, L = 1.8998, m = 2.1262,  = 0.0500.
unconstrained optimization problem, can also be solved And the corresponding dexterity is 15.8167.
using a direct search algorithm available from the Matlab
function: fminsearch. The method uses the concept of a It is noted that the new dexterity value is lower than the
simplex, which is a polytope of N+1 vertexes in N one found with n = 100 using the same initial guess
dimensions: a line segment in one dimension, a triangle (which was 17.4036) whereas the optimal parameters are
in two dimensions, a tetrahedron in three‐dimensional practically the same. The lower value for the dexterity is
space and so forth. The simplex algorithm is a direct expected due to use of a finer grid in area calculations.
search optimization method that begins at a starting Having the same optimal parameters indicates that
vertex and moves along the edges of the polytope until it changing the precision did not affect the optimal values
reaches the vertex of the optimum solution.
www.intechopen.com
for the parameters found by the direct search simplex [6] N. Simaan and M. Shoham, “Singularity Analysis of
algorithm but only improves the corresponding dexterity. a Class of Composite Serial In‐Parallel Robots,”
IEEE Transactions on Robotics and Automation, Vol.
5. Conclusions 17, No. 3, 2001, pp. 301‐311.
[7] M.Z. Huang and J.‐L. Thebert “A Study of
The problem and solution of an optimal design for a 2‐ Workspace and Singularity Characteristics for
DOF planar parallel manipulator were presented in this Design of 3‐DoF Planar Parallel Robots,”
paper. Using numerical simulations, the complete International Journal of Advanced Manufacturing
workspace and singularity were evaluated first. Then Technology, Vol. 51, Nos. 5‐8, 2010, pp. 789‐797.
using a dexterity index which quantifies the overall [8] Y. Lou, D. Zhang, and Z. Li, “Optimal Design of a
interplay between the workspace and the singularity, the Parallel Manipulator for Maximum Effective
design optimization were carried out two ways, first by Regular Workspace,” Proc. IEEE/RSJ International
the method of parametric variations and then the simplex Conference on Intelligent Robots System, Alberta,
direct search algorithm. An optimal design was obtained 2005, pp.795‐800.
and validated as a result. Moreover, it is found that [9] K. Miller, “Optimal Design and Modeling of Spatial
parametric variations could provide additional insights Parallel Manipulators,” International Journal of
into the interactions between the geometric parameters Robotics Research, Vol. 23, 2004, pp. 127‐140.
explicitly, which insight could also serve to inform and [10] X. J. Liu, “Optimal Kinematic Design of a Three
guide the automatic optimization process toward finding Translational DoFs Parallel Manipulator,” Robotica,
the global optimum. All results were presented Vol. 24, 2006, pp. 239‐250.
graphically to facilitate visualization. It is noted that the [11] S. Stan, V. Maties, R. Balan, and C. Lapusan,
work presented here dealt only with design optimization “Genetic Algorithms to Optimal Design of a 3 DOF
related to robot geometry; it did not explicitly address Parallel Robot,” Proc. 2008 IEEE International
other kinematic aspects such as singularity avoidance in Conference on Automation, Quality and Testing,
path planning. Robotics, Vol. 2, pp. 365‐370.
[12] F. Hao, J.‐P. Merlet, “Multicriteria Optimal Design
6. Acknowledgements of Parallel Manipulators Based on Interval
Analysis,” Mechanism and Machine Theory, Vol. 40,
Acknowledgment is due to SLT Henri Nicolas of the 2005, pp. 151‐171.
French Military Academy, St. Cyr, for his contributions in
generating the simulation results.

7. References

[1] G.R. Pennock and D.J. Kassner, “The Workspace of
a General Geometry Planar Three‐Degree‐of‐
Freedom Platform‐Type Manipulator,” ASME J. of
Mechanical Design, Vol. 115, June 1993, pp. 269‐276.
[2] J‐P. Merlet, C. Gosselin, and N. Mouly, “Workspace
of Planar Parallel Manipulators,” Mechanism and
Machine Theory, Vol. 33, 1998, pp. 7‐20.
[3] F. Gao, X‐J Liu, and X. Chen, “The Relationships
Between the Shapes of the Workspaces and the Link
Lengths of 3‐DOF Symmetrical Planar Parallel
Manipulators,” Mechanism and Machine Theory, Vol.
36, 2001, pp. 205‐220.
[4] J‐P. Merlet, “Singular Configurations of Parallel
Manipulators and Grassmann Geometry,” The
International Journal of Robotics Research, Vol. 8, No. 5,
1989, pp. 45‐56.
[5] L. Notash, “Uncertainty Configurations of Parallel
Manipulators,” Mechanism and Machine Theory, Vol.
33, No. ½, 1988, pp. 123‐138.

www.intechopen.com
View publication stats

Design of A Planar Parallel Robot For Optimal Workspace and Dexterity

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Design of A Planar Parallel Robot For Optimal Workspace and Dexterity

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.